Enhancing dynamic stability performance of hybrid ac/dc power grids using decentralised model predictive control

Abstract

High voltage direct current (HVDC) networks are transforming ac power grids to hybrid ac/dc power grids; hence hybrid ac/dc power grids are facing new stability challenges. In this paper, a decentralised control scheme based on model predictive control (MPC) is proposed to enhance the stability of hybrid ac/dc power grids under disturbances. The proposed scheme controls the dc voltage, frequency, active and reactive power simultaneously with minimum deviations from steady-state values. A communication network is used to share the sampled grid data with the local MPC of each voltage source converter (VSC). Hence, the central controller can be avoided. Moreover, the active power-sharing between converters can be pre-designed. Thus, fast and accurate active power control can be realised. Finally, the proposed control scheme is validated using a five-terminal VSCHVDC test system with MATLAB. The VSCHVDC network with the proposed control scheme shows a smaller deviation and a faster response compared with the conventional droop control scheme and centralised MPC scheme under small disturbances. Under large power variations, the proposed control scheme assists to reach the steady-state rapidly with a minimum impact on the entire power grid. The proposed control scheme has also proved its robustness under communication delays and outages.